setiastrosuitepro 1.6.7__py3-none-any.whl

setiastro/saspro/sfcc.py

@@ -0,0 +1,1530 @@
+# sfcc.py
+# SASpro Spectral Flux Color Calibration (SFCC) — "current view" integration
+# - Expects a "view adapter" you provide that exposes:
+#       get_rgb_image() -> np.ndarray (H,W,3), uint8 or float32 in [0,1]
+#       get_metadata() -> dict  (optional; may return {})
+#       get_header() -> astropy.io.fits.Header or dict (optional but needed for WCS features)
+#       set_rgb_image(img: np.ndarray, metadata: dict | None = None, step_name: str | None = None) -> None
+#   If your adapter names differ, tweak _get_img_meta/_get_header/_push_image below (they already try a few fallbacks).
+#
+# - Call open_sfcc(view_adapter, sasp_data_path) to show the dialog.
+
+from __future__ import annotations
+
+import os
+import re
+import cv2
+import math
+import time
+from datetime import datetime
+from typing import List, Tuple, Optional
+
+import numpy as np
+import pandas as pd
+
+# ── SciPy bits
+from scipy.interpolate import RBFInterpolator, interp1d
+from scipy.signal import medfilt
+
+# ── Astropy / Astroquery
+from astropy.io import fits
+from astropy.wcs import WCS
+import astropy.units as u
+from astropy.coordinates import SkyCoord
+from astroquery.simbad import Simbad
+
+# ── SEP (Source Extractor)
+import sep
+
+# ── Matplotlib backend for Qt
+from matplotlib.figure import Figure
+from matplotlib import pyplot as plt
+from matplotlib.backends.backend_qt5agg import FigureCanvasQTAgg as FigureCanvas
+
+from PyQt6.QtCore import (Qt, QPoint, QRect, QMimeData, QSettings, QByteArray,
+                          QDataStream, QIODevice, QEvent, QStandardPaths)
+from PyQt6.QtGui import (QAction, QDrag, QIcon, QMouseEvent, QPixmap, QKeyEvent)
+from PyQt6.QtWidgets import (QToolBar, QWidget, QToolButton, QMenu, QApplication, QVBoxLayout, QHBoxLayout, QComboBox, QGroupBox, QGridLayout, QDoubleSpinBox, QSpinBox,
+                             QInputDialog, QMessageBox, QDialog, QFileDialog,
+    QFormLayout, QDialogButtonBox, QDoubleSpinBox, QCheckBox, QLabel, QRubberBand, QRadioButton, QMainWindow, QPushButton)
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Utilities
+# ──────────────────────────────────────────────────────────────────────────────
+
+# --- Debug/guards -----------------------------------------------------
+def _debug_probe_channels(img: np.ndarray, label="input"):
+    assert img.ndim == 3 and img.shape[2] == 3, f"[SFCC] {label}: not RGB"
+    f = img.astype(np.float32) / (255.0 if img.dtype == np.uint8 else 1.0)
+    means = [float(f[...,i].mean()) for i in range(3)]
+    stds  = [float(f[...,i].std())  for i in range(3)]
+    rg = float(np.corrcoef(f[...,0].ravel(), f[...,1].ravel())[0,1])
+    rb = float(np.corrcoef(f[...,0].ravel(), f[...,2].ravel())[0,1])
+    gb = float(np.corrcoef(f[...,1].ravel(), f[...,2].ravel())[0,1])
+    print(f"[SFCC] {label}: mean={means}, std={stds}, corr(R,G)={rg:.5f}, corr(R,B)={rb:.5f}, corr(G,B)={gb:.5f}")
+    return rg, rb, gb
+
+def _maybe_bgr_to_rgb(img: np.ndarray) -> np.ndarray:
+    # Heuristic: if channel-2 is consistently brightest in highlights and ch-0 the dimmest → likely BGR.
+    f = img.astype(np.float32) / (255.0 if img.dtype == np.uint8 else 1.0)
+    lum = np.mean(f, axis=2)
+    thr = np.quantile(lum, 0.95)
+    m0 = f[...,0][lum >= thr].mean() if np.any(lum >= thr) else f[...,0].mean()
+    m1 = f[...,1][lum >= thr].mean() if np.any(lum >= thr) else f[...,1].mean()
+    m2 = f[...,2][lum >= thr].mean() if np.any(lum >= thr) else f[...,2].mean()
+    if (m2 > m1 >= m0) and (m2 - m0 > 0.02):
+        print("[SFCC] Heuristic suggests BGR input → converting to RGB")
+        return img[..., ::-1]
+    return img
+
+def _ensure_angstrom(wl: np.ndarray) -> np.ndarray:
+    """If wavelengths look like nm (≈300–1100), convert to Å."""
+    med = float(np.median(wl))
+    return wl * 10.0 if 250.0 <= med <= 2000.0 else wl
+
+
+def pickles_match_for_simbad(simbad_sp: str, available_extnames: List[str]) -> List[str]:
+    sp = simbad_sp.strip().upper()
+    if not sp:
+        return []
+    m = re.match(r"^([OBAFGKMLT])(\d?)(I{1,3}|IV|V)?", sp)
+    if not m:
+        return []
+    letter_class = m.group(1)
+    digit_part   = m.group(2)
+    lum_part     = m.group(3)
+    subclass = int(digit_part) if digit_part != "" else None
+
+    def parse_pickles_extname(ext: str):
+        ext = ext.strip().upper()
+        m2 = re.match(r"^([OBAFGKMLT])(\d+)(I{1,3}|IV|V)$", ext)
+        if not m2:
+            return None, None, None
+        return m2.group(1), int(m2.group(2)), m2.group(3)
+
+    parsed_templates = []
+    for ext in available_extnames:
+        l2, d2, L2 = parse_pickles_extname(ext)
+        if l2 is not None:
+            parsed_templates.append((ext, l2, d2, L2))
+
+    # Exact
+    if subclass is not None and lum_part is not None:
+        target = f"{letter_class}{subclass}{lum_part}"
+        if target in available_extnames:
+            return [target]
+
+    # Same letter (+same lum if we have it)
+    same_letter_and_lum = []
+    same_letter_any_lum = []
+    for (ext, l2, d2, L2) in parsed_templates:
+        if l2 != letter_class:
+            continue
+        if lum_part is not None and L2 == lum_part:
+            same_letter_and_lum.append((ext, d2))
+        else:
+            same_letter_any_lum.append((ext, d2))
+
+    def pick_nearest(candidates: List[Tuple[str, int]], target: int) -> List[str]:
+        if not candidates or target is None:
+            return []
+        arr = np.abs(np.array([d for _, d in candidates]) - target)
+        mind = np.min(arr)
+        return [candidates[i][0] for i in np.where(arr == mind)[0]]
+
+    if subclass is not None and lum_part is not None:
+        if same_letter_and_lum:
+            return pick_nearest(same_letter_and_lum, subclass)
+        if same_letter_any_lum:
+            return pick_nearest(same_letter_any_lum, subclass)
+
+    if subclass is not None and lum_part is None:
+        if same_letter_any_lum:
+            return pick_nearest(same_letter_any_lum, subclass)
+
+    if subclass is None and lum_part is None:
+        return sorted([ext for (ext, l2, _, _) in parsed_templates if l2 == letter_class])
+
+    if subclass is None and lum_part is not None:
+        cands = [ (ext, d2) for (ext, l2, d2, L2) in parsed_templates if l2 == letter_class and L2 == lum_part ]
+        if cands:
+            return sorted([ext for (ext, _) in cands])
+        return sorted([ext for (ext, l2, _, _) in parsed_templates if l2 == letter_class])
+
+    return []
+
+
+def compute_gradient_map(sources, delta_flux, shape, method="poly2"):
+    H, W = shape
+    xs, ys = sources[:, 0], sources[:, 1]
+
+    if method == "poly2":
+        A = np.vstack([np.ones_like(xs), xs, ys, xs**2, xs*ys, ys**2]).T
+        coeffs, *_ = np.linalg.lstsq(A, delta_flux, rcond=None)
+        YY, XX = np.mgrid[0:H, 0:W]
+        return (coeffs[0] + coeffs[1]*XX + coeffs[2]*YY
+                + coeffs[3]*XX**2 + coeffs[4]*XX*YY + coeffs[5]*YY**2)
+
+    elif method == "poly3":
+        A = np.vstack([
+            np.ones_like(xs), xs, ys,
+            xs**2, xs*ys, ys**2,
+            xs**3, xs**2*ys, xs*ys**2, ys**3
+        ]).T
+        coeffs, *_ = np.linalg.lstsq(A, delta_flux, rcond=None)
+        YY, XX = np.mgrid[0:H, 0:W]
+        return (coeffs[0] + coeffs[1]*XX + coeffs[2]*YY
+                + coeffs[3]*XX**2 + coeffs[4]*XX*YY + coeffs[5]*YY**2
+                + coeffs[6]*XX**3 + coeffs[7]*XX**2*YY + coeffs[8]*XX*YY**2 + coeffs[9]*YY**3)
+
+    elif method == "rbf":
+        pts = np.vstack([xs, ys]).T
+        rbfi = RBFInterpolator(pts, delta_flux, kernel="thin_plate_spline", smoothing=1.0)
+        YY, XX = np.mgrid[0:H, 0:W]
+        grid_pts = np.vstack([XX.ravel(), YY.ravel()]).T
+        return rbfi(grid_pts).reshape(H, W)
+
+    else:
+        raise ValueError("method must be one of 'poly2','poly3','rbf'")
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Simple responses viewer (unchanged core logic; useful for diagnostics)
+# ──────────────────────────────────────────────────────────────────────────────
+class SaspViewer(QMainWindow):
+    def __init__(self, sasp_data_path: str, user_custom_path: str):
+        super().__init__()
+        self.setWindowTitle(self.tr("SASP Viewer (Pickles + RGB Responses)"))
+
+        self.base_hdul   = fits.open(sasp_data_path,   mode="readonly", memmap=False)
+        self.custom_hdul = fits.open(user_custom_path, mode="readonly", memmap=False)
+
+        self.pickles_templates = []
+        self.filter_list       = []
+        self.sensor_list       = []
+        for hdul in (self.custom_hdul, self.base_hdul):
+            for hdu in hdul:
+                if not isinstance(hdu, fits.BinTableHDU): continue
+                c = hdu.header.get("CTYPE","").upper()
+                e = hdu.header.get("EXTNAME","")
+                if c == "SED":    self.pickles_templates.append(e)
+                elif c == "FILTER": self.filter_list.append(e)
+                elif c == "SENSOR": self.sensor_list.append(e)
+
+        for lst in (self.pickles_templates, self.filter_list, self.sensor_list):
+            lst.sort()
+        self.rgb_filter_choices = ["(None)"] + self.filter_list
+
+        central = QWidget(); self.setCentralWidget(central)
+        vbox = QVBoxLayout(); central.setLayout(vbox)
+
+        row = QHBoxLayout(); vbox.addLayout(row)
+        row.addWidget(QLabel(self.tr("Star Template:")))
+        self.star_combo = QComboBox(); self.star_combo.addItems(self.pickles_templates); row.addWidget(self.star_combo)
+        row.addWidget(QLabel(self.tr("R-Filter:")))
+        self.r_filter_combo = QComboBox(); self.r_filter_combo.addItems(self.rgb_filter_choices); row.addWidget(self.r_filter_combo)
+        row.addWidget(QLabel(self.tr("G-Filter:")))
+        self.g_filter_combo = QComboBox(); self.g_filter_combo.addItems(self.rgb_filter_choices); row.addWidget(self.g_filter_combo)
+        row.addWidget(QLabel(self.tr("B-Filter:")))
+        self.b_filter_combo = QComboBox(); self.b_filter_combo.addItems(self.rgb_filter_choices); row.addWidget(self.b_filter_combo)
+
+        row2 = QHBoxLayout(); vbox.addLayout(row2)
+        row2.addWidget(QLabel(self.tr("LP/Cut Filter1:")))
+        self.lp_filter_combo = QComboBox(); self.lp_filter_combo.addItems(self.rgb_filter_choices); row2.addWidget(self.lp_filter_combo)
+        row2.addWidget(QLabel(self.tr("LP/Cut Filter2:")))
+        self.lp_filter_combo2 = QComboBox(); self.lp_filter_combo2.addItems(self.rgb_filter_choices); row2.addWidget(self.lp_filter_combo2)
+        row2.addSpacing(20); row2.addWidget(QLabel(self.tr("Sensor (QE):")))
+        self.sens_combo = QComboBox(); self.sens_combo.addItems(self.sensor_list); row2.addWidget(self.sens_combo)
+
+        self.plot_btn = QPushButton(self.tr("Plot")); self.plot_btn.clicked.connect(self.update_plot); row.addWidget(self.plot_btn)
+
+        self.figure = Figure(figsize=(9, 6)); self.canvas = FigureCanvas(self.figure); vbox.addWidget(self.canvas)
+        self.update_plot()
+
+    def closeEvent(self, event):
+        self.base_hdul.close(); self.custom_hdul.close()
+        super().closeEvent(event)
+
+    def load_any(self, extname, field):
+        for hdul in (self.custom_hdul, self.base_hdul):
+            if extname in hdul:
+                return hdul[extname].data[field].astype(float)
+        raise KeyError(f"Extension '{extname}' not found")
+
+    def update_plot(self):
+        star_ext = self.star_combo.currentText()
+        r_filt   = self.r_filter_combo.currentText()
+        g_filt   = self.g_filter_combo.currentText()
+        b_filt   = self.b_filter_combo.currentText()
+        sens_ext = self.sens_combo.currentText()
+        lp_ext1  = self.lp_filter_combo.currentText()
+        lp_ext2  = self.lp_filter_combo2.currentText()
+
+        wl_star = self.load_any(star_ext, "WAVELENGTH")
+        fl_star = self.load_any(star_ext, "FLUX")
+        wl_sens = self.load_any(sens_ext, "WAVELENGTH")
+        qe_sens = self.load_any(sens_ext, "THROUGHPUT")
+
+        wl_min, wl_max = 1150.0, 10620.0
+        common_wl = np.arange(wl_min, wl_max + 1.0, 1.0)
+
+        sed_interp  = interp1d(wl_star, fl_star, kind="linear", bounds_error=False, fill_value=0.0)
+        sens_interp = interp1d(wl_sens, qe_sens, kind="linear", bounds_error=False, fill_value=0.0)
+        fl_common   = sed_interp(common_wl)
+        sens_common = sens_interp(common_wl)
+
+        rgb_data = {}
+        for color, filt_name in (("red", r_filt), ("green", g_filt), ("blue", b_filt)):
+            if filt_name == "(None)":
+                rgb_data[color] = None; continue
+
+            wl_filt = self.load_any(filt_name, "WAVELENGTH")
+            tr_filt = self.load_any(filt_name, "THROUGHPUT")
+            filt_common = interp1d(wl_filt, tr_filt, bounds_error=False, fill_value=0.0)(common_wl)
+
+            def lp_curve(ext):
+                if ext == "(None)": return np.ones_like(common_wl)
+                wl_lp = self.load_any(ext, "WAVELENGTH"); tr_lp = self.load_any(ext, "THROUGHPUT")
+                return interp1d(wl_lp, tr_lp, bounds_error=False, fill_value=0.0)(common_wl)
+
+            T_LP = lp_curve(lp_ext1) * lp_curve(lp_ext2)
+            T_sys = filt_common * sens_common * T_LP
+            resp  = fl_common * T_sys
+
+            rgb_data[color] = {"filter_name": filt_name, "T_sys": T_sys, "response": resp}
+
+        mag_texts = []
+        if "A0V" in self.pickles_templates:
+            wl_veg = self.load_any("A0V", "WAVELENGTH")
+            fl_veg = self.load_any("A0V", "FLUX")
+            fl_veg_c = interp1d(wl_veg, fl_veg, kind="linear", bounds_error=False, fill_value=0.0)(common_wl)
+            for color in ("red","green","blue"):
+                data = rgb_data[color]
+                if data is not None:
+                    S_star = np.trapezoid(data["response"], x=common_wl)
+                    S_veg  = np.trapezoid(fl_veg_c * data["T_sys"], x=common_wl)
+                    if S_veg>0 and S_star>0:
+                        mag = -2.5 * np.log10(S_star / S_veg)
+                        mag_texts.append(f"{color[0].upper()}→{data['filter_name']}: {mag:.2f}")
+                    else:
+                        mag_texts.append(f"{color[0].upper()}→{data['filter_name']}: N/A")
+        title_text = " | ".join(mag_texts) if mag_texts else self.tr("No channels selected")
+
+        self.figure.clf()
+        ax1 = self.figure.add_subplot(111)
+        ax1.plot(common_wl, fl_common, color="black", linewidth=1, label=f"{star_ext} SED")
+        for color, data in rgb_data.items():
+            if data is not None:
+                ax1.plot(common_wl, data["response"], color="gold", linewidth=1.5, label=self.tr("{0} Response").format(color.upper()))
+        ax1.set_xlim(wl_min, wl_max); ax1.set_xlabel(self.tr("Wavelength (Å)"))
+        ax1.set_ylabel(self.tr("Flux (erg s⁻¹ cm⁻² Å⁻¹)"), color="black"); ax1.tick_params(axis="y", labelcolor="black")
+
+        ax2 = ax1.twinx()
+        ax2.set_ylabel(self.tr("Relative Throughput"), color="red"); ax2.tick_params(axis="y", labelcolor="red"); ax2.set_ylim(0.0, 1.0)
+        if rgb_data["red"] is not None:   ax2.plot(common_wl, rgb_data["red"]["T_sys"],   color="red",   linestyle="--", linewidth=1, label=self.tr("R filter×QE"))
+        if rgb_data["green"] is not None: ax2.plot(common_wl, rgb_data["green"]["T_sys"], color="green", linestyle="--", linewidth=1, label=self.tr("G filter×QE"))
+        if rgb_data["blue"] is not None:  ax2.plot(common_wl, rgb_data["blue"]["T_sys"],  color="blue",  linestyle="--", linewidth=1, label=self.tr("B filter×QE"))
+
+        ax1.grid(True, which="both", linestyle="--", alpha=0.3); self.figure.suptitle(title_text, fontsize=10)
+        lines1, labels1 = ax1.get_legend_handles_labels(); lines2, labels2 = ax2.get_legend_handles_labels()
+        ax1.legend(lines1 + lines2, labels1 + labels2, loc="upper right")
+        self.canvas.draw()
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# SFCC Dialog (rewired for "current view")
+# ──────────────────────────────────────────────────────────────────────────────
+class SFCCDialog(QDialog):
+    """
+    Spectral Flux Color Calibration dialog, adapted for SASpro's current view.
+    Pass a 'view' adapter providing:
+       - get_rgb_image(), set_rgb_image(...)
+       - get_metadata()  [optional]
+       - get_header()    [preferred for WCS; else we look in metadata]
+    """
+    def __init__(self, doc_manager, sasp_data_path, parent=None):
+        super().__init__(parent)
+        self.setWindowTitle(self.tr("Spectral Flux Color Calibration"))
+        self.setWindowFlag(Qt.WindowType.Window, True)
+        self.setWindowModality(Qt.WindowModality.NonModal)
+        self.setModal(False)
+        try:
+            self.setAttribute(Qt.WidgetAttribute.WA_DeleteOnClose, True)
+        except Exception:
+            pass  # older PyQt6 versions        
+        self.setMinimumSize(800, 600)
+
+        self.doc_manager = doc_manager
+        self.sasp_data_path = sasp_data_path
+        self.user_custom_path = self._ensure_user_custom_fits()
+        self.current_image = None
+        self.current_header = None
+        self.orientation_label = QLabel(self.tr("Orientation: N/A"))
+        self.sasp_viewer_window = None
+        self.main_win = parent
+
+        # user custom file init … (unchanged)
+        # ...
+        self._reload_hdu_lists()
+        self.star_list = []
+        self._build_ui()
+        self.load_settings()
+
+        # persist combobox choices
+        self.r_filter_combo.currentIndexChanged.connect(self.save_r_filter_setting)
+        self.g_filter_combo.currentIndexChanged.connect(self.save_g_filter_setting)
+        self.b_filter_combo.currentIndexChanged.connect(self.save_b_filter_setting)
+        self.lp_filter_combo.currentIndexChanged.connect(self.save_lp_setting)
+        self.lp_filter_combo2.currentIndexChanged.connect(self.save_lp2_setting)
+        self.sens_combo.currentIndexChanged.connect(self.save_sensor_setting)
+        self.star_combo.currentIndexChanged.connect(self.save_star_setting)
+        self.finished.connect(lambda *_: self._cleanup())
+
+        self.grad_method = "poly3"
+        self.grad_method_combo.currentTextChanged.connect(lambda m: setattr(self, "grad_method", m))
+
+    # ── View plumbing ───────────────────────────────────────────────────
+    def _get_active_image_and_header(self):
+        doc = self.doc_manager.get_active_document()
+        if doc is None:
+            return None, None, None
+
+        img = doc.image
+        meta = doc.metadata or {}
+
+        # Prefer the normalized WCS header if present, then fall back
+        hdr = (
+            meta.get("wcs_header") or
+            meta.get("original_header") or
+            meta.get("header")
+        )
+
+        return img, hdr, meta
+
+    
+    def _get_img_meta(self) -> Tuple[Optional[np.ndarray], dict]:
+        """Try a few common shapes to obtain image + metadata from the view."""
+        meta = {}
+        img = None
+        if hasattr(self.view, "get_image_and_metadata"):
+            try:
+                img, meta = self.view.get_image_and_metadata()
+            except Exception:
+                pass
+        if img is None and hasattr(self.view, "get_rgb_image"):
+            img = self.view.get_rgb_image()
+        if not meta and hasattr(self.view, "get_metadata"):
+            try:
+                meta = self.view.get_metadata() or {}
+            except Exception:
+                meta = {}
+        return img, (meta or {})
+
+    def _get_header(self):
+        header = None
+        if hasattr(self.view, "get_header"):
+            try:
+                header = self.view.get_header()
+            except Exception:
+                header = None
+        if header is None:
+            # fall back to metadata
+            _, meta = self._get_img_meta()
+            header = meta.get("original_header") or meta.get("header")
+        return header
+
+    def _push_image(self, img: np.ndarray, meta: Optional[dict], step_name: str):
+        """Send image back to the same current view."""
+        if hasattr(self.view, "set_rgb_image"):
+            self.view.set_rgb_image(img, meta or {}, step_name)
+        elif hasattr(self.view, "set_image"):
+            self.view.set_image(img, meta or {}, step_name=step_name)
+        elif hasattr(self.view, "update_image"):
+            self.view.update_image(img, meta or {}, step_name=step_name)
+        else:
+            # As a last resort, try attribute assignment (for custom apps)
+            if hasattr(self.view, "image"):
+                self.view.image = img
+            if hasattr(self.view, "metadata"):
+                self.view.metadata = meta or {}
+
+    # ── File prep ───────────────────────────────────────────────────────
+
+    def _ensure_user_custom_fits(self) -> str:
+        app_data = QStandardPaths.writableLocation(QStandardPaths.StandardLocation.AppDataLocation)
+        os.makedirs(app_data, exist_ok=True)
+        path = os.path.join(app_data, "usercustomcurves.fits")
+        if not os.path.exists(path):
+            fits.HDUList([fits.PrimaryHDU()]).writeto(path)
+        return path
+
+    # ── UI ──────────────────────────────────────────────────────────────
+
+    def _build_ui(self):
+        layout = QVBoxLayout(self)
+
+        row1 = QHBoxLayout(); layout.addLayout(row1)
+        self.fetch_stars_btn = QPushButton(self.tr("Step 1: Fetch Stars from Current View"))
+        f = self.fetch_stars_btn.font(); f.setBold(True); self.fetch_stars_btn.setFont(f)
+        self.fetch_stars_btn.clicked.connect(self.fetch_stars)
+        row1.addWidget(self.fetch_stars_btn)
+
+        self.open_sasp_btn = QPushButton(self.tr("Open SASP Viewer"))
+        self.open_sasp_btn.clicked.connect(self.open_sasp_viewer)
+        row1.addWidget(self.open_sasp_btn)
+
+        row1.addSpacing(20)
+        row1.addWidget(QLabel(self.tr("Select White Reference:")))
+        self.star_combo = QComboBox()
+        self.star_combo.addItem(self.tr("Vega (A0V)"), userData="A0V")
+        for sed in getattr(self, "sed_list", []):
+            if sed.upper() == "A0V": continue
+            self.star_combo.addItem(sed, userData=sed)
+        row1.addWidget(self.star_combo)
+        idx_g2v = self.star_combo.findData("G2V")
+        if idx_g2v >= 0: self.star_combo.setCurrentIndex(idx_g2v)
+
+        row2 = QHBoxLayout(); layout.addLayout(row2)
+        row2.addWidget(QLabel(self.tr("R Filter:")))
+        self.r_filter_combo = QComboBox(); self.r_filter_combo.addItem("(None)"); self.r_filter_combo.addItems(self.filter_list); row2.addWidget(self.r_filter_combo)
+        row2.addSpacing(20); row2.addWidget(QLabel(self.tr("G Filter:")))
+        self.g_filter_combo = QComboBox(); self.g_filter_combo.addItem("(None)"); self.g_filter_combo.addItems(self.filter_list); row2.addWidget(self.g_filter_combo)
+        row2.addSpacing(20); row2.addWidget(QLabel(self.tr("B Filter:")))
+        self.b_filter_combo = QComboBox(); self.b_filter_combo.addItem("(None)"); self.b_filter_combo.addItems(self.filter_list); row2.addWidget(self.b_filter_combo)
+
+        row3 = QHBoxLayout(); layout.addLayout(row3)
+        row3.addStretch()
+        row3.addWidget(QLabel(self.tr("Sensor (QE):")))
+        self.sens_combo = QComboBox(); self.sens_combo.addItem("(None)"); self.sens_combo.addItems(self.sensor_list); row3.addWidget(self.sens_combo)
+        row3.addSpacing(20); row3.addWidget(QLabel(self.tr("LP/Cut Filter1:")))
+        self.lp_filter_combo = QComboBox(); self.lp_filter_combo.addItem("(None)"); self.lp_filter_combo.addItems(self.filter_list); row3.addWidget(self.lp_filter_combo)
+        row3.addSpacing(20); row3.addWidget(QLabel(self.tr("LP/Cut Filter2:")))
+        self.lp_filter_combo2 = QComboBox(); self.lp_filter_combo2.addItem("(None)"); self.lp_filter_combo2.addItems(self.filter_list); row3.addWidget(self.lp_filter_combo2)
+        row3.addStretch()
+
+        row4 = QHBoxLayout(); layout.addLayout(row4)
+        self.run_spcc_btn = QPushButton(self.tr("Step 2: Run Color Calibration"))
+        f2 = self.run_spcc_btn.font(); f2.setBold(True); self.run_spcc_btn.setFont(f2)
+        self.run_spcc_btn.clicked.connect(self.run_spcc)
+        row4.addWidget(self.run_spcc_btn)
+
+        self.neutralize_chk = QCheckBox(self.tr("Background Neutralization")); self.neutralize_chk.setChecked(True); row4.addWidget(self.neutralize_chk)
+
+        self.run_grad_btn = QPushButton(self.tr("Run Gradient Extraction (Beta)"))
+        f3 = self.run_grad_btn.font(); f3.setBold(True); self.run_grad_btn.setFont(f3)
+        self.run_grad_btn.clicked.connect(self.run_gradient_extraction)
+        row4.addWidget(self.run_grad_btn)
+
+        self.grad_method_combo = QComboBox(); self.grad_method_combo.addItems(["poly2","poly3","rbf"]); self.grad_method_combo.setCurrentText("poly3")
+        row4.addWidget(self.grad_method_combo)
+
+        row4.addSpacing(15)
+        row4.addWidget(QLabel(self.tr("Star detect σ:")))
+        self.sep_thr_spin = QSpinBox()
+        self.sep_thr_spin.setRange(2, 50)        # should be enough
+        self.sep_thr_spin.setValue(5)            # our current hardcoded value
+        self.sep_thr_spin.valueChanged.connect(self.save_sep_threshold_setting)
+        row4.addWidget(self.sep_thr_spin)
+
+        row4.addStretch()
+        self.add_curve_btn = QPushButton(self.tr("Add Custom Filter/Sensor Curve…"))
+        self.add_curve_btn.clicked.connect(self.add_custom_curve); row4.addWidget(self.add_curve_btn)
+        self.remove_curve_btn = QPushButton(self.tr("Remove Filter/Sensor Curve…"))
+        self.remove_curve_btn.clicked.connect(self.remove_custom_curve); row4.addWidget(self.remove_curve_btn)
+        row4.addStretch()
+        self.close_btn = QPushButton(self.tr("Close")); self.close_btn.clicked.connect(self.reject); row4.addWidget(self.close_btn)
+
+        self.count_label = QLabel(""); layout.addWidget(self.count_label)
+
+        self.figure = Figure(figsize=(6, 4)); self.canvas = FigureCanvas(self.figure); self.canvas.setVisible(False); layout.addWidget(self.canvas, stretch=1)
+        self.reset_btn = QPushButton(self.tr("Reset View/Close")); self.reset_btn.clicked.connect(self.reject); layout.addWidget(self.reset_btn)
+
+        # hide gradient controls by default (enable if you like)
+        self.run_grad_btn.hide(); self.grad_method_combo.hide()
+        layout.addWidget(self.orientation_label)
+
+    # ── Settings helpers ────────────────────────────────────────────────
+
+    def _reload_hdu_lists(self):
+        self.sed_list = []
+        with fits.open(self.sasp_data_path, mode="readonly", memmap=False) as base:
+            for hdu in base:
+                if isinstance(hdu, fits.BinTableHDU) and hdu.header.get("CTYPE","").upper()=="SED":
+                    self.sed_list.append(hdu.header["EXTNAME"])
+
+        self.filter_list = []; self.sensor_list = []
+        for path in (self.sasp_data_path, self.user_custom_path):
+            with fits.open(path, mode="readonly", memmap=False) as hdul:
+                for hdu in hdul:
+                    if not isinstance(hdu, fits.BinTableHDU): continue
+                    c = hdu.header.get("CTYPE","").upper(); e = hdu.header.get("EXTNAME","")
+                    if c=="FILTER": self.filter_list.append(e)
+                    elif c=="SENSOR": self.sensor_list.append(e)
+        self.sed_list.sort(); self.filter_list.sort(); self.sensor_list.sort()
+
+    def load_settings(self):
+        s = QSettings()
+        def apply(cb, key):
+            val = s.value(key, "")
+            if val:
+                idx = cb.findText(val)
+                if idx != -1:
+                    cb.setCurrentIndex(idx)
+
+        # existing stuff...
+        saved_star = QSettings().value("SFCC/WhiteReference", "")
+        if saved_star:
+            idx = self.star_combo.findText(saved_star)
+            if idx != -1:
+                self.star_combo.setCurrentIndex(idx)
+
+        apply(self.r_filter_combo, "SFCC/RFilter")
+        apply(self.g_filter_combo, "SFCC/GFilter")
+        apply(self.b_filter_combo, "SFCC/BFilter")
+        apply(self.sens_combo,     "SFCC/Sensor")
+        apply(self.lp_filter_combo,  "SFCC/LPFilter")
+        apply(self.lp_filter_combo2, "SFCC/LPFilter2")
+
+        # 👇 NEW: load SEP/star-detect threshold
+        sep_thr = int(s.value("SFCC/SEPThreshold", 5))
+        if hasattr(self, "sep_thr_spin"):
+            self.sep_thr_spin.setValue(sep_thr)
+    def save_sep_threshold_setting(self, v: int):
+        QSettings().setValue("SFCC/SEPThreshold", int(v))
+
+    def save_lp_setting(self, _):  QSettings().setValue("SFCC/LPFilter", self.lp_filter_combo.currentText())
+    def save_lp2_setting(self, _): QSettings().setValue("SFCC/LPFilter2", self.lp_filter_combo2.currentText())
+    def save_star_setting(self, _): QSettings().setValue("SFCC/WhiteReference", self.star_combo.currentText())
+    def save_r_filter_setting(self, _): QSettings().setValue("SFCC/RFilter", self.r_filter_combo.currentText())
+    def save_g_filter_setting(self, _): QSettings().setValue("SFCC/GFilter", self.g_filter_combo.currentText())
+    def save_b_filter_setting(self, _): QSettings().setValue("SFCC/BFilter", self.b_filter_combo.currentText())
+    def save_sensor_setting(self, _):   QSettings().setValue("SFCC/Sensor", self.sens_combo.currentText())
+
+    # ── Curve utilities ─────────────────────────────────────────────────
+
+    def interpolate_bad_points(self, wl, tr):
+        tr = tr.copy()
+        bad = (tr < 0.0) | (tr > 1.0)
+        good = ~bad
+        if not np.any(bad): return tr, np.array([], dtype=int)
+        if np.sum(good) < 2: raise RuntimeError("Not enough valid points to interpolate anomalies.")
+        tr_corr = tr.copy()
+        tr_corr[bad] = np.interp(wl[bad], wl[good], tr[good])
+        return tr_corr, np.where(bad)[0]
+
+    def smooth_curve(self, tr, window_size=5):
+        return medfilt(tr, kernel_size=window_size)
+
+    def get_calibration_points(self, rgb_img: np.ndarray):
+        print("\nClick three calibration points: BL (λmin,0), BR (λmax,0), TL (λmin,1)")
+        fig, ax = plt.subplots(figsize=(8, 5)); ax.imshow(rgb_img); ax.set_title(self.tr("Click 3 points, then close"))
+        pts = plt.ginput(3, timeout=-1); plt.close(fig)
+        if len(pts) != 3: raise RuntimeError(self.tr("Need exactly three clicks for calibration."))
+        return pts[0], pts[1], pts[2]
+
+    def build_transforms(self, px_bl, py_bl, px_br, py_br, px_tl, py_tl, λ_min, λ_max, resp_min, resp_max):
+        nm_per_px = (λ_max - λ_min) / (px_br - px_bl)
+        resp_per_px = (resp_max - resp_min) / (py_bl - py_tl)
+        def px_to_λ(px):  return λ_min + (px - px_bl) * nm_per_px
+        def py_to_resp(py): return resp_max - (py - py_tl) * resp_per_px
+        return px_to_λ, py_to_resp
+
+    def extract_curve(self, gray_img, λ_mapper, resp_mapper, λ_min, λ_max, threshold=50):
+        H, W = gray_img.shape
+        data = []
+        for px in range(W):
+            col = gray_img[:, px]
+            py_min = int(np.argmin(col)); val_min = int(col[py_min])
+            if val_min < threshold:
+                lam = λ_mapper(px)
+                if λ_min <= lam <= λ_max:
+                    data.append((lam, resp_mapper(py_min)))
+        if not data:
+            raise RuntimeError("No dark pixels found; raise threshold or adjust clicks.")
+        df = (pd.DataFrame(data, columns=["wavelength_nm", "response"])
+                .sort_values("wavelength_nm").reset_index(drop=True))
+        df = df[(df["wavelength_nm"] >= λ_min) & (df["wavelength_nm"] <= λ_max)].copy()
+        if df["wavelength_nm"].iloc[0] > λ_min:
+            df = pd.concat([pd.DataFrame([[λ_min, 0.0]], columns=["wavelength_nm", "response"]), df], ignore_index=True)
+        if df["wavelength_nm"].iloc[-1] < λ_max:
+            df = pd.concat([df, pd.DataFrame([[λ_max, 0.0]], columns=["wavelength_nm", "response"])], ignore_index=True)
+        return df.sort_values("wavelength_nm").reset_index(drop=True)
+
+    def _query_name_channel(self):
+        name_str, ok1 = QInputDialog.getText(self, self.tr("Curve Name"), self.tr("Enter curve name (EXTNAME):"))
+        if not (ok1 and name_str.strip()): return False, None, None
+        extname = name_str.strip().upper().replace(" ", "_")
+        ch_str, ok2 = QInputDialog.getText(self, self.tr("Channel"), self.tr("Enter channel (R,G,B or Q for sensor):"))
+        if not (ok2 and ch_str.strip()): return False, None, None
+        return True, extname, ch_str.strip().upper()
+
+    def _append_curve_hdu(self, wl_ang, tr_final, extname, ctype, origin):
+        col_wl = fits.Column(name="WAVELENGTH", format="E", unit="Angstrom", array=wl_ang.astype(np.float32))
+        col_tr = fits.Column(name="THROUGHPUT", format="E", unit="REL",      array=tr_final.astype(np.float32))
+        new_hdu = fits.BinTableHDU.from_columns([col_wl, col_tr])
+        new_hdu.header["EXTNAME"] = extname
+        new_hdu.header["CTYPE"]   = ctype
+        new_hdu.header["ORIGIN"]  = origin
+        with fits.open(self.user_custom_path, mode="update", memmap=False) as hdul:
+            hdul.append(new_hdu); hdul.flush()
+
+    def add_custom_curve(self):
+        msg = QMessageBox(self); msg.setWindowTitle(self.tr("Add Custom Curve")); msg.setText(self.tr("Choose how to add the curve:"))
+        csv_btn = msg.addButton(self.tr("Import CSV"), QMessageBox.ButtonRole.AcceptRole)
+        img_btn = msg.addButton(self.tr("Digitize Image"), QMessageBox.ButtonRole.AcceptRole)
+        cancel_btn = msg.addButton(QMessageBox.StandardButton.Cancel)
+        msg.exec()
+        if msg.clickedButton() == csv_btn: self._import_curve_from_csv()
+        elif msg.clickedButton() == img_btn: self._digitize_curve_from_image()
+
+    def _import_curve_from_csv(self):
+        csv_path, _ = QFileDialog.getOpenFileName(self, self.tr("Select 2-column CSV (λ_nm, response)"), "", "CSV Files (*.csv);;All Files (*)")
+        if not csv_path: return
+        try:
+            df = (pd.read_csv(csv_path, comment="#", header=None).iloc[:, :2].dropna())
+            df.columns = ["wavelength_nm","response"]
+            wl_nm = df["wavelength_nm"].astype(float).to_numpy(); tp = df["response"].astype(float).to_numpy()
+        except ValueError:
+            try:
+                df = (pd.read_csv(csv_path, comment="#", header=0).iloc[:, :2].dropna())
+                df.columns = ["wavelength_nm","response"]
+                wl_nm = df["wavelength_nm"].astype(float).to_numpy(); tp = df["response"].astype(float).to_numpy()
+            except Exception as e2:
+                QMessageBox.critical(self, self.tr("CSV Error"), self.tr("Could not read CSV:\n{0}").format(e2)); return
+        except Exception as e:
+            QMessageBox.critical(self, self.tr("CSV Error"), self.tr("Could not read CSV:\n{0}").format(e)); return
+
+        ok, extname_base, channel_val = self._query_name_channel()
+        if not ok: return
+        wl_ang = (wl_nm * 10.0).astype(np.float32); tr_final = tp.astype(np.float32)
+        self._append_curve_hdu(wl_ang, tr_final, extname_base, "SENSOR" if channel_val=="Q" else "FILTER", f"CSV:{os.path.basename(csv_path)}")
+        self._reload_hdu_lists(); self.refresh_filter_sensor_lists()
+        QMessageBox.information(self, self.tr("Done"), self.tr("CSV curve '{0}' added.").format(extname_base))
+
+    def _digitize_curve_from_image(self):
+        img_path_str, _ = QFileDialog.getOpenFileName(self, self.tr("Select Curve Image to Digitize"), "", "Images (*.png *.jpg *.jpeg *.bmp);;All Files (*)")
+        if not img_path_str: return
+        img_filename = os.path.basename(img_path_str)
+        try:
+            bgr = cv2.imread(img_path_str)
+            if bgr is None: raise RuntimeError(f"cv2.imread returned None for '{img_path_str}'")
+            rgb_img = cv2.cvtColor(bgr, cv2.COLOR_BGR2RGB); gray_img = cv2.cvtColor(bgr, cv2.COLOR_BGR2GRAY)
+        except Exception as e:
+            QMessageBox.critical(self, self.tr("Error"), self.tr("Could not load image:\n{0}").format(e)); return
+
+        try:
+            (px_bl, py_bl), (px_br, py_br), (px_tl, py_tl) = self.get_calibration_points(rgb_img)
+        except Exception as e:
+            QMessageBox.critical(self, self.tr("Digitization Error"), str(e)); return
+
+        λ_min_str, ok1 = QInputDialog.getText(self, self.tr("λ_min"), self.tr("Enter λ_min (in nm):"))
+        λ_max_str, ok2 = QInputDialog.getText(self, self.tr("λ_max"), self.tr("Enter λ_max (in nm):"))
+        if not (ok1 and ok2 and λ_min_str.strip() and λ_max_str.strip()): return
+        try:
+            λ_min = float(λ_min_str); λ_max = float(λ_max_str)
+        except ValueError:
+            QMessageBox.critical(self, self.tr("Input Error"), self.tr("λ_min and λ_max must be numbers.")); return
+
+        ok, extname_base, channel_val = self._query_name_channel()
+        if not ok: return
+
+        px_to_λ, py_to_resp = self.build_transforms(px_bl, py_bl, px_br, py_br, px_tl, py_tl, λ_min, λ_max, 0.0, 1.0)
+        try:
+            df_curve = self.extract_curve(gray_img, px_to_λ, py_to_resp, λ_min, λ_max, threshold=50)
+        except Exception as e:
+            QMessageBox.critical(self, self.tr("Extraction Error"), str(e)); return
+
+        df_curve["wl_int"] = df_curve["wavelength_nm"].round().astype(int)
+        grp = (df_curve.groupby("wl_int")["response"].median().reset_index().sort_values("wl_int"))
+        wl = grp["wl_int"].to_numpy(dtype=int); tr = grp["response"].to_numpy(dtype=float)
+
+        try:
+            tr_corr, _ = self.interpolate_bad_points(wl, tr)
+        except Exception as e:
+            QMessageBox.critical(self, self.tr("Interpolation Error"), str(e)); return
+
+        tr_smoothed = self.smooth_curve(tr_corr, window_size=5)
+        wl_ang = (wl.astype(float) * 10.0).astype(np.float32); tr_final = tr_smoothed.astype(np.float32)
+        self._append_curve_hdu(wl_ang, tr_final, extname_base, "SENSOR" if channel_val=="Q" else "FILTER", f"UserDefined:{img_filename}")
+        self._reload_hdu_lists(); self.refresh_filter_sensor_lists()
+        QMessageBox.information(self, self.tr("Done"), self.tr("Added curve '{0}'.").format(extname_base))
+
+    def remove_custom_curve(self):
+        all_curves = self.filter_list + self.sensor_list
+        if not all_curves:
+            QMessageBox.information(self, self.tr("Remove Curve"), self.tr("No custom curves to remove.")); return
+        curve, ok = QInputDialog.getItem(self, self.tr("Remove Curve"), self.tr("Select a FILTER or SENSOR curve to delete:"), all_curves, 0, False)
+        if not ok or not curve: return
+        reply = QMessageBox.question(self, self.tr("Confirm Deletion"), self.tr("Delete '{0}'?").format(curve), QMessageBox.StandardButton.Yes | QMessageBox.StandardButton.No)
+        if reply != QMessageBox.StandardButton.Yes: return
+
+        temp_path = self.user_custom_path + ".tmp"
+        try:
+            with fits.open(self.user_custom_path, memmap=False) as old_hdul:
+                new_hdus = []
+                for hdu in old_hdul:
+                    if hdu is old_hdul[0]:
+                        new_hdus.append(hdu.copy())
+                    else:
+                        if hdu.header.get("EXTNAME") != curve:
+                            new_hdus.append(hdu.copy())
+            fits.HDUList(new_hdus).writeto(temp_path, overwrite=True)
+            os.replace(temp_path, self.user_custom_path)
+        except Exception as e:
+            if os.path.exists(temp_path): os.remove(temp_path)
+            QMessageBox.critical(self, "Write Error", f"Could not remove curve:\n{e}"); return
+
+        self._reload_hdu_lists(); self.refresh_filter_sensor_lists()
+        QMessageBox.information(self, self.tr("Removed"), self.tr("Deleted curve '{0}'.").format(curve))
+
+    def refresh_filter_sensor_lists(self):
+        self._reload_hdu_lists()
+        current_r = self.r_filter_combo.currentText()
+        current_g = self.g_filter_combo.currentText()
+        current_b = self.b_filter_combo.currentText()
+        current_s = self.sens_combo.currentText()
+        current_lp  = self.lp_filter_combo.currentText()
+        current_lp2 = self.lp_filter_combo2.currentText()
+
+        for cb, lst, prev in [
+            (self.r_filter_combo, self.filter_list, current_r),
+            (self.g_filter_combo, self.filter_list, current_g),
+            (self.b_filter_combo, self.filter_list, current_b),
+        ]:
+            cb.clear(); cb.addItem("(None)"); cb.addItems(lst)
+            idx = cb.findText(prev);  cb.setCurrentIndex(idx if idx != -1 else 0)
+
+        for cb, prev in [(self.lp_filter_combo, current_lp), (self.lp_filter_combo2, current_lp2)]:
+            cb.clear(); cb.addItem("(None)"); cb.addItems(self.filter_list)
+            idx = cb.findText(prev);  cb.setCurrentIndex(idx if idx != -1 else 0)
+
+        self.sens_combo.clear(); self.sens_combo.addItem("(None)"); self.sens_combo.addItems(self.sensor_list)
+        idx = self.sens_combo.findText(current_s); self.sens_combo.setCurrentIndex(idx if idx != -1 else 0)
+
+    # ── WCS utilities ──────────────────────────────────────────────────
+    def initialize_wcs_from_header(self, header):
+        if header is None:
+            print("No FITS header available; cannot build WCS.")
+            return
+        try:
+            hdr = header.copy()
+
+            # --- normalize deprecated keywords ---
+            if "RADECSYS" in hdr and "RADESYS" not in hdr:
+                radesys_val = str(hdr["RADECSYS"]).strip()
+                hdr["RADESYS"] = radesys_val
+                try:
+                    del hdr["RADECSYS"]
+                except Exception:
+                    pass
+
+                alt_letters = {
+                    k[-1]
+                    for k in hdr.keys()
+                    if re.match(r"^CTYPE[12][A-Z]$", k)
+                }
+                for a in alt_letters:
+                    key = f"RADESYS{a}"
+                    if key not in hdr:
+                        hdr[key] = radesys_val
+
+            if "EPOCH" in hdr and "EQUINOX" not in hdr:
+                hdr["EQUINOX"] = hdr["EPOCH"]
+                try:
+                    del hdr["EPOCH"]
+                except Exception:
+                    pass
+
+            # IMPORTANT: use the normalized hdr, not the original header
+            self.wcs = WCS(hdr, naxis=2, relax=True)
+
+            psm = self.wcs.pixel_scale_matrix
+            self.pixscale = np.hypot(psm[0, 0], psm[1, 0]) * 3600.0
+            self.center_ra, self.center_dec = self.wcs.wcs.crval
+            self.wcs_header = self.wcs.to_header(relax=True)
+
+            # Orientation from normalized header
+            if "CROTA2" in hdr:
+                try:
+                    self.orientation = float(hdr["CROTA2"])
+                except Exception:
+                    self.orientation = None
+            else:
+                self.orientation = self.calculate_orientation(hdr)
+
+            if self.orientation is not None:
+                self.orientation_label.setText(f"Orientation: {self.orientation:.2f}°")
+            else:
+                self.orientation_label.setText("Orientation: N/A")
+
+        except Exception as e:
+            print("WCS initialization error:\n", e)
+
+
+    def calculate_orientation(self, header):
+        try:
+            cd1_1 = float(header.get("CD1_1", 0.0))
+            cd1_2 = float(header.get("CD1_2", 0.0))
+            return math.degrees(math.atan2(cd1_2, cd1_1))
+        except Exception:
+            return None
+
+    def calculate_ra_dec_from_pixel(self, x, y):
+        if not hasattr(self, "wcs"): return None, None
+        return self.wcs.all_pix2world(x, y, 0)
+
+    # ── Background neutralization ───────────────────────────────────────
+
+    def _neutralize_background(self, rgb_img: np.ndarray, patch_size: int = 50) -> np.ndarray:
+        img = rgb_img.copy()
+        h, w = img.shape[:2]
+        ph, pw = h // patch_size, w // patch_size
+        min_sum, best_med = np.inf, None
+        for i in range(patch_size):
+            for j in range(patch_size):
+                y0, x0 = i * ph, j * pw
+                patch = img[y0:min(y0+ph, h), x0:min(x0+pw, w), :]
+                med   = np.median(patch, axis=(0, 1))
+                s     = med.sum()
+                if s < min_sum:
+                    min_sum, best_med = s, med
+        if best_med is None:
+            return img
+        target = float(best_med.mean()); eps = 1e-8
+        for c in range(3):
+            diff = float(best_med[c] - target)
+            if abs(diff) < eps: continue
+            img[..., c] = np.clip((img[..., c] - diff) / (1.0 - diff), 0.0, 1.0)
+        return img
+
+    # ── SIMBAD/Star fetch ──────────────────────────────────────────────
+
+    def fetch_stars(self):
+        # 0) Grab current image + header from the active document
+        img, hdr, _meta = self._get_active_image_and_header()
+        self.current_image = img
+        self.current_header = hdr
+
+        if self.current_header is None or self.current_image is None:
+            QMessageBox.warning(self, "No Plate Solution",
+                                "Please plate-solve the active document first.")
+            return
+
+        # Pickles templates list (once)
+        if not hasattr(self, "pickles_templates"):
+            self.pickles_templates = []
+            for p in (self.user_custom_path, self.sasp_data_path):
+                try:
+                    with fits.open(p) as hd:
+                        for hdu in hd:
+                            if (isinstance(hdu, fits.BinTableHDU)
+                                    and hdu.header.get("CTYPE", "").upper() == "SED"):
+                                extname = hdu.header.get("EXTNAME", None)
+                                if extname and extname not in self.pickles_templates:
+                                    self.pickles_templates.append(extname)
+                except Exception as e:
+                    print(f"[fetch_stars] Could not load Pickles templates from {p}: {e}")
+
+        # Build WCS
+        try:
+            self.initialize_wcs_from_header(self.current_header)
+        except Exception:
+            QMessageBox.critical(self, "WCS Error", "Could not build a 2D WCS from header."); return
+
+        H, W = self.current_image.shape[:2]
+        pix = np.array([[W/2, H/2], [0,0], [W,0], [0,H], [W,H]])
+        try:
+            sky = self.wcs.all_pix2world(pix, 0)
+        except Exception as e:
+            QMessageBox.critical(self, "WCS Conversion Error", str(e)); return
+        center_sky  = SkyCoord(ra=sky[0,0]*u.deg, dec=sky[0,1]*u.deg, frame="icrs")
+        corners_sky = SkyCoord(ra=sky[1:,0]*u.deg, dec=sky[1:,1]*u.deg, frame="icrs")
+        radius_deg  = center_sky.separation(corners_sky).max().deg
+
+        # Simbad fields
+        Simbad.reset_votable_fields()
+        for attempt in range(1, 6):
+            try:
+                Simbad.add_votable_fields('sp', 'flux(B)', 'flux(V)', 'flux(R)')
+                break
+            except Exception:
+                QApplication.processEvents()
+                time.sleep(1.2)
+        Simbad.ROW_LIMIT = 10000
+
+        for attempt in range(1, 6):
+            try:
+                result = Simbad.query_region(center_sky, radius=radius_deg * u.deg)
+                break
+            except Exception as e:
+                self.count_label.setText(f"Attempt {attempt}/5 to query SIMBAD…")
+                QApplication.processEvents(); time.sleep(1.2)
+                result = None
+        if result is None or len(result) == 0:
+            QMessageBox.information(self, "No Stars", "SIMBAD returned zero objects in that region.")
+            self.star_list = []; self.star_combo.clear(); self.star_combo.addItem("Vega (A0V)", userData="A0V"); return
+
+        def infer_letter(bv):
+            if bv is None or (isinstance(bv, float) and np.isnan(bv)): return None
+            if   bv < 0.00: return "B"
+            elif bv < 0.30: return "A"
+            elif bv < 0.58: return "F"
+            elif bv < 0.81: return "G"
+            elif bv < 1.40: return "K"
+            elif bv > 1.40: return "M"
+            else: return "U"
+
+        self.star_list = []; templates_for_hist = []
+        for row in result:
+            raw_sp = row['sp_type']
+            bmag, vmag, rmag = row['B'], row['V'], row['R']
+            ra_deg, dec_deg  = float(row['ra']), float(row['dec'])
+            try:
+                sc = SkyCoord(ra=ra_deg*u.deg, dec=dec_deg*u.deg, frame="icrs")
+            except Exception:
+                continue
+
+            def _unmask_num(x):
+                try:
+                    if x is None or np.ma.isMaskedArray(x) and np.ma.is_masked(x):
+                        return None
+                    return float(x)
+                except Exception:
+                    return None
+
+            # inside your SIMBAD row loop:
+            bmag = _unmask_num(row['B'])
+            vmag = _unmask_num(row['V'])
+
+            sp_clean = None
+            if raw_sp and str(raw_sp).strip():
+                sp = str(raw_sp).strip().upper()
+                if not (sp.startswith("SN") or sp.startswith("KA")):
+                    sp_clean = sp
+            elif bmag is not None and vmag is not None:
+                bv = bmag - vmag
+                sp_clean = infer_letter(bv)
+            if not sp_clean: continue
+
+            match_list = pickles_match_for_simbad(sp_clean, self.pickles_templates)
+            best_template = match_list[0] if match_list else None
+            xpix, ypix = self.wcs.all_world2pix(sc.ra.deg, sc.dec.deg, 0)
+            if 0 <= xpix < W and 0 <= ypix < H:
+                self.star_list.append({
+                    "ra": sc.ra.deg, "dec": sc.dec.deg, "sp_clean": sp_clean,
+                    "pickles_match": best_template, "x": xpix, "y": ypix,
+                    "Bmag": float(bmag) if bmag else None,
+                    "Vmag": float(vmag) if vmag else None,
+                    "Rmag": float(rmag) if rmag else None,
+                })
+                if best_template is not None: templates_for_hist.append(best_template)
+
+        self.figure.clf()
+        if templates_for_hist:
+            uniq, cnt = np.unique(templates_for_hist, return_counts=True)
+            types_str = ", ".join(uniq)
+            self.count_label.setText(f"Found {len(templates_for_hist)} stars; templates: {types_str}")
+            ax = self.figure.add_subplot(111)
+            ax.bar(uniq, cnt, edgecolor="black")
+            ax.set_xlabel("Spectral Type"); ax.set_ylabel("Count"); ax.set_title("Spectral Distribution")
+            ax.tick_params(axis='x', rotation=90); ax.grid(axis="y", linestyle="--", alpha=0.3)
+            self.canvas.setVisible(True); self.canvas.draw()
+        else:
+            self.count_label.setText("Found 0 stars with Pickles matches.")
+            self.canvas.setVisible(False); self.canvas.draw()
+
+    # ── Core SFCC ───────────────────────────────────────────────────────
+
+    def run_spcc(self):
+        ref_sed_name = self.star_combo.currentData()
+        r_filt = self.r_filter_combo.currentText()
+        g_filt = self.g_filter_combo.currentText()
+        b_filt = self.b_filter_combo.currentText()
+        sens_name = self.sens_combo.currentText()
+        lp_filt  = self.lp_filter_combo.currentText()
+        lp_filt2 = self.lp_filter_combo2.currentText()
+
+        if not ref_sed_name:
+            QMessageBox.warning(self, "Error", "Select a reference spectral type (e.g. A0V)."); return
+        if r_filt == "(None)" and g_filt == "(None)" and b_filt == "(None)":
+            QMessageBox.warning(self, "Error", "Pick at least one of R, G or B filters."); return
+        if sens_name == "(None)":
+            QMessageBox.warning(self, "Error", "Select a sensor QE curve."); return
+
+        # -- Step 1A: get active image as float32 in [0..1]
+        doc = self.doc_manager.get_active_document()
+        if doc is None or doc.image is None:
+            QMessageBox.critical(self, "Error", "No active document.")
+            return
+
+        img = doc.image
+        H, W = img.shape[:2]
+        if img.ndim != 3 or img.shape[2] != 3:
+            QMessageBox.critical(self, "Error", "Active document must be RGB (3 channels).")
+            return
+
+        if img.dtype == np.uint8:
+            base = img.astype(np.float32) / 255.0
+        else:
+            base = img.astype(np.float32, copy=True)
+
+        # pedestal removal
+        base = np.clip(base - np.min(base, axis=(0,1)), 0.0, None)
+        # light neutralization
+        base = self._neutralize_background(base, patch_size=10)
+
+        # SEP on grayscale
+        gray = np.mean(base, axis=2)
+        
+        bkg = sep.Background(gray)
+        data_sub = gray - bkg.back()
+        err = bkg.globalrms
+
+        # 👇 get user threshold (default 5.0)
+        if hasattr(self, "sep_thr_spin"):
+            sep_sigma = float(self.sep_thr_spin.value())
+        else:
+            sep_sigma = 5.0
+        self.count_label.setText(f"Detecting stars (SEP σ={sep_sigma:.1f})…"); QApplication.processEvents()
+        sources = sep.extract(data_sub, sep_sigma, err=err)
+
+        MAX_SOURCES = 300_000
+        if sources.size > MAX_SOURCES:
+            QMessageBox.warning(
+                self,
+                "Too many detections",
+                f"SEP found {sources.size:,} sources with σ={sep_sigma:.1f}.\n"
+                f"Increase the threshold and rerun SFCC."
+            )
+            return
+
+        if sources.size == 0:
+            QMessageBox.critical(self, "SEP Error", "SEP found no sources."); return
+        r_fluxrad, _ = sep.flux_radius(gray, sources["x"], sources["y"], 2.0*sources["a"], 0.5, normflux=sources["flux"], subpix=5)
+        mask = (r_fluxrad > .2) & (r_fluxrad <= 10); sources = sources[mask]
+        if sources.size == 0:
+            QMessageBox.critical(self, "SEP Error", "All SEP detections rejected by radius filter."); return
+
+        if not getattr(self, "star_list", None):
+            QMessageBox.warning(self, "Error", "Fetch Stars (with WCS) before running SFCC."); return
+
+        raw_matches = []
+        for i, star in enumerate(self.star_list):
+            dx = sources["x"] - star["x"]; dy = sources["y"] - star["y"]
+            j = np.argmin(dx*dx + dy*dy)
+            if (dx[j]**2 + dy[j]**2) < 3.0**2:
+                xi, yi = int(round(sources["x"][j])), int(round(sources["y"][j]))
+                if 0 <= xi < W and 0 <= yi < H:
+                    raw_matches.append({"sim_index": i, "template": star.get("pickles_match") or star["sp_clean"], "x_pix": xi, "y_pix": yi})
+        if not raw_matches:
+            QMessageBox.warning(self, "No Matches", "No SIMBAD star matched to SEP detections."); return
+
+        wl_min, wl_max = 3000, 11000
+        wl_grid = np.arange(wl_min, wl_max+1)
+
+        def load_curve(ext):
+            for p in (self.user_custom_path, self.sasp_data_path):
+                with fits.open(p) as hd:
+                    if ext in hd:
+                        d = hd[ext].data
+                        wl = _ensure_angstrom(d["WAVELENGTH"].astype(float))
+                        tp = d["THROUGHPUT"].astype(float)
+                        return wl, tp
+            raise KeyError(f"Curve '{ext}' not found")
+
+        def load_sed(ext):
+            for p in (self.user_custom_path, self.sasp_data_path):
+                with fits.open(p) as hd:
+                    if ext in hd:
+                        d = hd[ext].data
+                        wl = _ensure_angstrom(d["WAVELENGTH"].astype(float))
+                        fl = d["FLUX"].astype(float)
+                        return wl, fl
+            raise KeyError(f"SED '{ext}' not found")
+
+        interp = lambda wl_o, tp_o: np.interp(wl_grid, wl_o, tp_o, left=0., right=0.)
+        T_R = interp(*load_curve(r_filt)) if r_filt!="(None)" else np.ones_like(wl_grid)
+        T_G = interp(*load_curve(g_filt)) if g_filt!="(None)" else np.ones_like(wl_grid)
+        T_B = interp(*load_curve(b_filt)) if b_filt!="(None)" else np.ones_like(wl_grid)
+        QE  = interp(*load_curve(sens_name)) if sens_name!="(None)" else np.ones_like(wl_grid)
+        LP1 = interp(*load_curve(lp_filt))   if lp_filt != "(None)"  else np.ones_like(wl_grid)
+        LP2 = interp(*load_curve(lp_filt2))  if lp_filt2!= "(None)"  else np.ones_like(wl_grid)
+        LP  = LP1 * LP2
+        T_sys_R, T_sys_G, T_sys_B = T_R*QE*LP, T_G*QE*LP, T_B*QE*LP
+
+        wl_ref, fl_ref = load_sed(ref_sed_name)
+        fr_i = np.interp(wl_grid, wl_ref, fl_ref, left=0., right=0.)
+        S_ref_R = np.trapezoid(fr_i * T_sys_R, x=wl_grid)
+        S_ref_G = np.trapezoid(fr_i * T_sys_G, x=wl_grid)
+        S_ref_B = np.trapezoid(fr_i * T_sys_B, x=wl_grid)
+
+        diag_meas_RG, diag_exp_RG = [], []
+        diag_meas_BG, diag_exp_BG = [], []
+        enriched = []
+
+        # --- Optimization: Pre-calculate integrals for unique templates ---
+        unique_simbad_types = set(m["template"] for m in raw_matches)
+        
+        # Map simbad_type -> pickles_template_name
+        simbad_to_pickles = {}
+        pickles_templates_needed = set()
+        
+        for sp in unique_simbad_types:
+            cands = pickles_match_for_simbad(sp, getattr(self, "pickles_templates", []))
+            if cands:
+                pickles_name = cands[0]
+                simbad_to_pickles[sp] = pickles_name
+                pickles_templates_needed.add(pickles_name)
+
+        # Pre-calc integrals for each unique Pickles template
+        # Cache structure: template_name -> (S_sr, S_sg, S_sb)
+        template_integrals = {}
+
+        # Cache for load_sed to avoid re-reading even across different calls if desired, 
+        # but here we just optimize the loop.
+        
+        for pname in pickles_templates_needed:
+            try:
+                wl_s, fl_s = load_sed(pname)
+                fs_i = np.interp(wl_grid, wl_s, fl_s, left=0., right=0.)
+                
+                S_sr = np.trapezoid(fs_i * T_sys_R, x=wl_grid)
+                S_sg = np.trapezoid(fs_i * T_sys_G, x=wl_grid)
+                S_sb = np.trapezoid(fs_i * T_sys_B, x=wl_grid)
+                
+                template_integrals[pname] = (S_sr, S_sg, S_sb)
+            except Exception as e:
+                print(f"[SFCC] Warning: failed to load/integrate template {pname}: {e}")
+
+        # --- Main Match Loop ---
+        for m in raw_matches:
+            xi, yi, sp = m["x_pix"], m["y_pix"], m["template"]
+            Rm = float(base[yi, xi, 0]); Gm = float(base[yi, xi, 1]); Bm = float(base[yi, xi, 2])
+            if Gm <= 0: continue
+
+            # 1. Resolve Simbad -> Pickles
+            pname = simbad_to_pickles.get(sp)
+            if not pname: continue
+            
+            # 2. Retrieve pre-calced integrals
+            integrals = template_integrals.get(pname)
+            if not integrals: continue
+            
+            S_sr, S_sg, S_sb = integrals
+            
+            if S_sg <= 0: continue
+
+            exp_RG = S_sr / S_sg; exp_BG = S_sb / S_sg
+            meas_RG = Rm / Gm;    meas_BG = Bm / Gm
+
+            diag_meas_RG.append(meas_RG); diag_exp_RG.append(exp_RG)
+            diag_meas_BG.append(meas_BG); diag_exp_BG.append(exp_BG)
+
+            enriched.append({
+                **m, "R_meas": Rm, "G_meas": Gm, "B_meas": Bm,
+                "S_star_R": S_sr, "S_star_G": S_sg, "S_star_B": S_sb,
+                "exp_RG": exp_RG, "exp_BG": exp_BG
+            })
+            
+        self._last_matched = enriched
+        diag_meas_RG = np.array(diag_meas_RG); diag_exp_RG = np.array(diag_exp_RG)
+        diag_meas_BG = np.array(diag_meas_BG); diag_exp_BG = np.array(diag_exp_BG)
+        if diag_meas_RG.size == 0 or diag_meas_BG.size == 0:
+            QMessageBox.information(self, "No Valid Stars", "No stars with valid measured vs expected ratios."); return
+        n_stars = diag_meas_RG.size
+
+        def rms_frac(pred, exp): return np.sqrt(np.mean(((pred/exp) - 1.0) ** 2))
+        slope_only = lambda x, m: m*x
+        affine     = lambda x, m, b: m*x + b
+        quad       = lambda x, a, b, c: a*x**2 + b*x + c
+
+        denR = np.sum(diag_meas_RG**2); denB = np.sum(diag_meas_BG**2)
+        mR_s = (np.sum(diag_meas_RG * diag_exp_RG) / denR) if denR > 0 else 1.0
+        mB_s = (np.sum(diag_meas_BG * diag_exp_BG) / denB) if denB > 0 else 1.0
+        rms_s = rms_frac(slope_only(diag_meas_RG, mR_s), diag_exp_RG) + rms_frac(slope_only(diag_meas_BG, mB_s), diag_exp_BG)
+
+        mR_a, bR_a = np.linalg.lstsq(np.vstack([diag_meas_RG, np.ones_like(diag_meas_RG)]).T, diag_exp_RG, rcond=None)[0]
+        mB_a, bB_a = np.linalg.lstsq(np.vstack([diag_meas_BG, np.ones_like(diag_meas_BG)]).T, diag_exp_BG, rcond=None)[0]
+        rms_a = rms_frac(affine(diag_meas_RG, mR_a, bR_a), diag_exp_RG) + rms_frac(affine(diag_meas_BG, mB_a, bB_a), diag_exp_BG)
+
+        aR_q, bR_q, cR_q = np.polyfit(diag_meas_RG, diag_exp_RG, 2)
+        aB_q, bB_q, cB_q = np.polyfit(diag_meas_BG, diag_exp_BG, 2)
+        rms_q = rms_frac(quad(diag_meas_RG, aR_q, bR_q, cR_q), diag_exp_RG) + rms_frac(quad(diag_meas_BG, aB_q, bB_q, cB_q), diag_exp_BG)
+
+        idx = np.argmin([rms_s, rms_a, rms_q])
+        if idx == 0: coeff_R, coeff_B, model_choice = (0, mR_s, 0), (0, mB_s, 0), "slope-only"
+        elif idx == 1: coeff_R, coeff_B, model_choice = (0, mR_a, bR_a), (0, mB_a, bB_a), "affine"
+        else: coeff_R, coeff_B, model_choice = (aR_q, bR_q, cR_q), (aB_q, bB_q, cB_q), "quadratic"
+
+        poly = lambda c, x: c[0]*x**2 + c[1]*x + c[2]
+        self.figure.clf()
+        #ax1 = self.figure.add_subplot(1, 3, 1); bins=20
+        #ax1.hist(diag_meas_RG, bins=bins, alpha=.65, label="meas R/G", color="firebrick", edgecolor="black")
+        #ax1.hist(diag_exp_RG,  bins=bins, alpha=.55, label="exp R/G",  color="salmon",   edgecolor="black")
+        #ax1.hist(diag_meas_BG, bins=bins, alpha=.65, label="meas B/G", color="royalblue", edgecolor="black")
+        #ax1.hist(diag_exp_BG,  bins=bins, alpha=.55, label="exp B/G",  color="lightskyblue", edgecolor="black")
+        #ax1.set_xlabel("Ratio (band / G)"); ax1.set_ylabel("Count"); ax1.set_title("Measured vs expected"); ax1.legend(fontsize=7, frameon=False)
+
+        res0_RG = (diag_meas_RG / diag_exp_RG) - 1.0
+        res0_BG = (diag_meas_BG / diag_exp_BG) - 1.0
+        res1_RG = (poly(coeff_R, diag_meas_RG) / diag_exp_RG) - 1.0
+        res1_BG = (poly(coeff_B, diag_meas_BG) / diag_exp_BG) - 1.0
+
+        ymin = np.min(np.concatenate([res0_RG, res0_BG])); ymax = np.max(np.concatenate([res0_RG, res0_BG]))
+        pad  = 0.05 * (ymax - ymin) if ymax > ymin else 0.02; y_lim = (ymin - pad, ymax + pad)
+        def shade(ax, yvals, color):
+            q1, q3 = np.percentile(yvals, [25,75]); ax.axhspan(q1, q3, color=color, alpha=.10, zorder=0)
+
+        ax2 = self.figure.add_subplot(1, 2, 1)
+        ax2.axhline(0, color="0.65", ls="--", lw=1); shade(ax2, res0_RG, "firebrick"); shade(ax2, res0_BG, "royalblue")
+        ax2.scatter(diag_exp_RG, res0_RG, c="firebrick",  marker="o", alpha=.7, label="R/G residual")
+        ax2.scatter(diag_exp_BG, res0_BG, c="royalblue", marker="s", alpha=.7, label="B/G residual")
+        ax2.set_ylim(*y_lim); ax2.set_xlabel("Expected (band/G)"); ax2.set_ylabel("Frac residual (meas/exp − 1)")
+        ax2.set_title("Residuals • BEFORE"); ax2.legend(frameon=False, fontsize=7, loc="lower right")
+
+        ax3 = self.figure.add_subplot(1, 2, 2)
+        ax3.axhline(0, color="0.65", ls="--", lw=1); shade(ax3, res1_RG, "firebrick"); shade(ax3, res1_BG, "royalblue")
+        ax3.scatter(diag_exp_RG, res1_RG, c="firebrick",  marker="o", alpha=.7)
+        ax3.scatter(diag_exp_BG, res1_BG, c="royalblue", marker="s", alpha=.7)
+        ax3.set_ylim(*y_lim); ax3.set_xlabel("Expected (band/G)"); ax3.set_ylabel("Frac residual (corrected/exp − 1)")
+        ax3.set_title("Residuals • AFTER")
+        self.canvas.setVisible(True); self.figure.tight_layout(w_pad=2.); self.canvas.draw()
+
+        self.count_label.setText("Applying SFCC color scales to image…"); QApplication.processEvents()
+        if img.dtype == np.uint8: img_float = img.astype(np.float32) / 255.0
+        else:                     img_float = img.astype(np.float32)
+
+        RG = img_float[..., 0] / np.maximum(img_float[..., 1], 1e-8)
+        BG = img_float[..., 2] / np.maximum(img_float[..., 1], 1e-8)
+        aR, bR, cR = coeff_R; aB, bB, cB = coeff_B
+        RG_corr = aR*RG**2 + bR*RG + cR
+        BG_corr = aB*BG**2 + bB*BG + cB
+        calibrated = img_float.copy()
+        calibrated[..., 0] = RG_corr * img_float[..., 1]
+        calibrated[..., 2] = BG_corr * img_float[..., 1]
+        calibrated = np.clip(calibrated, 0, 1)
+
+        if self.neutralize_chk.isChecked():
+            calibrated = self._neutralize_background(calibrated, patch_size=10)
+
+        if img.dtype == np.uint8:
+            calibrated = (np.clip(calibrated, 0, 1) * 255.0).astype(np.uint8)
+        else:
+            calibrated = np.clip(calibrated, 0, 1).astype(np.float32)
+
+        new_meta = dict(doc.metadata or {})
+        new_meta.update({
+            "SFCC_applied": True,
+            "SFCC_timestamp": datetime.now().isoformat(),
+            "SFCC_model": model_choice,
+            "SFCC_coeff_R": [float(v) for v in coeff_R],
+            "SFCC_coeff_B": [float(v) for v in coeff_B],
+        })
+
+        self.doc_manager.update_active_document(
+            calibrated,
+            metadata=new_meta,
+            step_name="SFCC Calibrated",
+            doc=doc,   # 👈 pin to the document we started from
+        )
+
+        self.count_label.setText(f"Applied SFCC color calibration using {n_stars} stars")
+        QApplication.processEvents()
+
+        def pretty(coeff): return coeff[0] + coeff[1] + coeff[2]
+        ratio_R, ratio_B = pretty(coeff_R), pretty(coeff_B)
+        QMessageBox.information(self, "SFCC Complete",
+                                f"Applied SFCC using {n_stars} stars\n"
+                                f"Model: {model_choice}\n"
+                                f"R ratio @ x=1: {ratio_R:.4f}\n"
+                                f"B ratio @ x=1: {ratio_B:.4f}\n"
+                                f"Background neutralisation: {'ON' if self.neutralize_chk.isChecked() else 'OFF'}")
+
+        self.current_image = calibrated  # keep for gradient step
+
+    # ── Chromatic gradient (optional) ──────────────────────────────────
+
+    def run_gradient_extraction(self):
+        if not getattr(self, "_last_matched", None):
+            QMessageBox.warning(self, "No Star Matches", "Run colour calibration first.")
+            return
+
+        doc = self.doc_manager.get_active_document()
+        if doc is None or doc.image is None:
+            QMessageBox.critical(self, "Error", "No active document.")
+            return
+
+        img = doc.image
+        if img.ndim != 3 or img.shape[2] != 3:
+            QMessageBox.critical(self, "Error", "Active document must be RGB.")
+            return
+
+        is_u8 = (img.dtype == np.uint8)
+        img_f = img.astype(np.float32) / (255.0 if is_u8 else 1.0)
+        H, W = img_f.shape[:2]
+
+        # Need star diagnostics from SPCC
+        if not hasattr(self, "_last_matched") or not self._last_matched:
+            QMessageBox.warning(self, "No Star Matches", "Run color calibration first."); return
+
+        down_fact = 4
+        Hs, Ws = H // down_fact, W // down_fact
+        small  = cv2.resize(img_f, (Ws, Hs), interpolation=cv2.INTER_AREA)
+
+        pts, dRG, dBG = [], [], []
+        eps, box = 1e-8, 3
+        for st in self._last_matched:
+            xs_full, ys_full = st["x_pix"], st["y_pix"]
+            xs, ys = xs_full / down_fact, ys_full / down_fact
+            xs_c, ys_c = int(round(xs)), int(round(ys))
+            if not (0 <= xs_c < Ws and 0 <= ys_c < Hs): continue
+            xsl = slice(max(0, xs_c-box), min(Ws, xs_c+box+1))
+            ysl = slice(max(0, ys_c-box), min(Hs, ys_c+box+1))
+            Rm = np.median(small[ysl, xsl, 0]); Gm = np.median(small[ysl, xsl, 1]); Bm = np.median(small[ysl, xsl, 2])
+            if Gm <= 0: continue
+            meas_RG = Rm / Gm; meas_BG = Bm / Gm
+            exp_RG, exp_BG = st["exp_RG"], st["exp_BG"]
+            if exp_RG is None or exp_BG is None: continue
+            dm_RG = -2.5 * np.log10((meas_RG+eps)/(exp_RG+eps))
+            dm_BG = -2.5 * np.log10((meas_BG+eps)/(exp_BG+eps))
+            pts.append([xs, ys]); dRG.append(dm_RG); dBG.append(dm_BG)
+
+        pts  = np.asarray(pts); dRG = np.asarray(dRG); dBG = np.asarray(dBG)
+        if pts.shape[0] < 5:
+            QMessageBox.warning(self, "Too Few Stars", "Need ≥5 stars after clipping."); return
+
+        def sclip(arr, p, s=2.5):
+            m, sd = np.median(arr), np.std(arr); keep = np.abs(arr-m) < s*sd
+            return p[keep], arr[keep]
+
+        ptsRG, dRG = sclip(dRG, pts); ptsBG, dBG = sclip(dBG, pts)
+
+        mode = getattr(self, "grad_method", "poly2")
+        bgRG_s = compute_gradient_map(ptsRG, dRG, (Hs, Ws), method=mode)
+        bgBG_s = compute_gradient_map(ptsBG, dBG, (Hs, Ws), method=mode)
+
+        for bg in (bgRG_s, bgBG_s):
+            bg -= np.median(bg)
+            peak = np.max(np.abs(bg))
+            if peak > 0.2: bg *= 0.2/peak
+
+        bgRG = cv2.resize(bgRG_s, (W, H), interpolation=cv2.INTER_CUBIC)
+        bgBG = cv2.resize(bgBG_s, (W, H), interpolation=cv2.INTER_CUBIC)
+
+        scale_R = 10**(-0.4*bgRG); scale_B = 10**(-0.4*bgBG)
+
+        self.figure.clf()
+        for i,(surf,lbl) in enumerate(((bgRG,"Δm R/G"),(bgBG,"Δm B/G"))):
+            ax  = self.figure.add_subplot(1,2,i+1)
+            im  = ax.imshow(surf, origin="lower", cmap="RdBu")
+            ax.set_title(lbl); self.figure.colorbar(im, ax=ax)
+        self.canvas.setVisible(True); self.figure.tight_layout(); self.canvas.draw()
+
+        corrected = img_f.copy()
+        corrected[...,0] = np.clip(corrected[...,0] / scale_R, 0, 1.0)
+        corrected[...,2] = np.clip(corrected[...,2] / scale_B, 0, 1.0)
+        corrected = np.clip(corrected, 0, 1)
+        if is_u8:
+            corrected = (corrected * 255.0).astype(np.uint8)
+        else:
+            corrected = corrected.astype(np.float32)
+
+        new_meta = dict(doc.metadata or {})
+        new_meta["ColourGradRemoved"] = True
+
+        self.doc_manager.update_active_document(
+            corrected,
+            metadata=new_meta,
+            step_name="Colour-Gradient (star spectra, ¼-res fit)",
+            doc=doc,   # 👈 same idea
+        )
+        self.count_label.setText("Chromatic gradient removed ✓")
+        QApplication.processEvents()
+
+    # ── Viewer, close ──────────────────────────────────────────────────
+
+    def open_sasp_viewer(self):
+        if self.sasp_viewer_window is not None:
+            if self.sasp_viewer_window.isVisible():
+                self.sasp_viewer_window.raise_()
+            else:
+                self.sasp_viewer_window.show()
+            return
+
+        self.sasp_viewer_window = SaspViewer(
+            sasp_data_path=self.sasp_data_path,
+            user_custom_path=self.user_custom_path
+        )
+        self.sasp_viewer_window.show()
+        self.sasp_viewer_window.destroyed.connect(self._on_sasp_closed)
+
+    def _cleanup(self):
+        # 1) Close/cleanup child window (SaspViewer)
+        try:
+            if getattr(self, "sasp_viewer_window", None) is not None:
+                try:
+                    self.sasp_viewer_window.destroyed.disconnect(self._on_sasp_closed)
+                except Exception:
+                    pass
+                try:
+                    self.sasp_viewer_window.close()
+                except Exception:
+                    pass
+                self.sasp_viewer_window = None
+        except Exception:
+            pass
+
+        # 2) Disconnect any long-lived external signals (add these if/when used)
+        # Example patterns:
+        try:
+            self.doc_manager.activeDocumentChanged.disconnect(self._on_active_doc_changed)
+        except Exception:
+            pass
+        try:
+            self.main_win.currentDocumentChanged.disconnect(self._on_active_doc_changed)
+        except Exception:
+            pass
+
+        # 3) Release large caches/refs (important since dialog may not be deleted)
+        try:
+            self.current_image = None
+            self.current_header = None
+            self.star_list = []
+            self._last_matched = []
+            if hasattr(self, "wcs"):
+                self.wcs = None
+            if hasattr(self, "wcs_header"):
+                self.wcs_header = None
+        except Exception:
+            pass
+
+        # 4) Matplotlib cleanup
+        try:
+            if getattr(self, "figure", None) is not None:
+                self.figure.clf()
+            if getattr(self, "canvas", None) is not None:
+                self.canvas.setVisible(False)
+                self.canvas.draw_idle()
+        except Exception:
+            pass
+
+
+    def _on_sasp_closed(self, _=None):
+        # Called when the SaspViewer window is destroyed
+        self.sasp_viewer_window = None
+        self._cleanup()
+
+    def closeEvent(self, event):
+        self._cleanup()
+        super().closeEvent(event)
+
+
+# ──────────────────────────────────────────────────────────────────────────────
+# Helper to open the dialog from your app
+# ──────────────────────────────────────────────────────────────────────────────
+
+def open_sfcc(doc_manager, sasp_data_path: str, parent=None) -> SFCCDialog:
+    dlg = SFCCDialog(doc_manager=doc_manager, sasp_data_path=sasp_data_path, parent=parent)
+    dlg.show()
+    return dlg